The 2020 US elections news coverage was extensive, with new pieces of information generated rapidly. This evolving scenario presented an opportunity to study the performance of search engines in a context in which they had to quickly process information as it was published. We analyze novelty, a measurement of new items that emerge in the top news search results, to compare the coverage and visibility of different topics. We conduct a longitudinal study of news results of five search engines collected in short-bursts (every 21 minutes) from two regions (Oregon, US and Frankfurt, Germany), starting on election day and lasting until one day after the announcement of Biden as the winner. We find more new items emerging for election related queries ("joe biden", "donald trump" and "us elections") compared to topical (e.g., "coronavirus") or stable (e.g., "holocaust") queries. We demonstrate differences across search engines and regions over time, and we highlight imbalances between candidate queries. When it comes to news search, search engines are responsible for such imbalances, either due to their algorithms or the set of news sources they rely on. We argue that such imbalances affect the visibility of political candidates in news searches during electoral periods.
Understanding the dynamics and reactions of cells from population snapshots is a major challenge in single-cell transcriptomics. Here, we present Geodesic Sinkhorn, a method for interpolating populations along a data manifold that leverages existing kernels developed for single-cell dimensionality reduction and visualization methods. Our Geodesic Sinkhorn method uses a heat-geodesic ground distance that, as compared to Euclidean ground distances, is more accurate for interpolating single-cell dynamics on a wide variety of datasets and significantly speeds up the computation for sparse kernels. We first apply Geodesic Sinkhorn to 10 single-cell transcriptomics time series interpolation datasets as a drop-in replacement for existing interpolation methods where it outperforms on all datasets, showing its effectiveness in modeling cell dynamics. Second, we show how to efficiently approximate the operator with polynomial kernels allowing us to improve scaling to large datasets. Finally, we define the conditional Wasserstein-average treatment effect and show how it can elucidate the treatment effect on single-cell populations on a drug screen.
This work studies the use of attention masking in transformer transducer based speech recognition for building a single configurable model for different deployment scenarios. We present a comprehensive set of experiments comparing fixed masking, where the same attention mask is applied at every frame, with chunked masking, where the attention mask for each frame is determined by chunk boundaries, in terms of recognition accuracy and latency. We then explore the use of variable masking, where the attention masks are sampled from a target distribution at training time, to build models that can work in different configurations. Finally, we investigate how a single configurable model can be used to perform both first pass streaming recognition and second pass acoustic rescoring. Experiments show that chunked masking achieves a better accuracy vs latency trade-off compared to fixed masking, both with and without FastEmit. We also show that variable masking improves the accuracy by up to 8% relative in the acoustic re-scoring scenario.
The size of deep neural networks has grown exponentially in recent years. Unfortunately, hardware devices have not kept pace with the rapidly increasing memory requirements. To cope with this, researchers have turned to techniques such as spilling and recomputation, which increase training time, or reduced precision and model pruning, which can affect model accuracy. We present OLLA, an algorithm that optimizes the lifetime and memory location of the tensors used to train neural networks. Our method reduces the memory usage of existing neural networks, without needing any modification to the models or their training procedures. We formulate the problem as a joint integer linear program (ILP). We present several techniques to simplify the encoding of the problem, and enable our approach to scale to the size of state-of-the-art neural networks using an off-the-shelf ILP solver. We experimentally demonstrate that OLLA only takes minutes if not seconds to allow the training of neural networks using one-third less memory on average.
Clinical adoption of personalized virtual heart simulations faces challenges in model personalization and expensive computation. While an ideal solution is an efficient neural surrogate that at the same time is personalized to an individual subject, the state-of-the-art is either concerned with personalizing an expensive simulation model, or learning an efficient yet generic surrogate. This paper presents a completely new concept to achieve personalized neural surrogates in a single coherent framework of meta-learning (metaPNS). Instead of learning a single neural surrogate, we pursue the process of learning a personalized neural surrogate using a small amount of context data from a subject, in a novel formulation of few-shot generative modeling underpinned by: 1) a set-conditioned neural surrogate for cardiac simulation that, conditioned on subject-specific context data, learns to generate query simulations not included in the context set, and 2) a meta-model of amortized variational inference that learns to condition the neural surrogate via simple feed-forward embedding of context data. As test time, metaPNS delivers a personalized neural surrogate by fast feed-forward embedding of a small and flexible number of data available from an individual, achieving -- for the first time -- personalization and surrogate construction for expensive simulations in one end-to-end learning framework. Synthetic and real-data experiments demonstrated that metaPNS was able to improve personalization and predictive accuracy in comparison to conventionally-optimized cardiac simulation models, at a fraction of computation.
We elucidate why an interval algorithm that computes the exact bounds on the amplitude and phase of the discrete Fourier transform can run in polynomial time. We address this question from a formal perspective to provide the mathematical foundations underpinning such an algorithm. We show that the procedure set out by the algorithm fully addresses the dependency problem of interval arithmetic, making it usable in a variety of applications involving the discrete Fourier transform. For example when analysing signals with poor precision, signals with missing data, and for automatic error propagation and verified computations.
Clinical Cohort Studies (CCS), such as randomized clinical trials, are a great source of documented clinical research. Ideally, a clinical expert inspects these articles for exploratory analysis ranging from drug discovery for evaluating the efficacy of existing drugs in tackling emerging diseases to the first test of newly developed drugs. However, more than 100 articles are published daily on a single prevalent disease like COVID-19 in PubMed. As a result, it can take days for a physician to find articles and extract relevant information. Can we develop a system to sift through the long list of these articles faster and document the crucial takeaways from each of these articles? In this work, we propose CCS Explorer, an end-to-end system for relevance prediction of sentences, extractive summarization, and patient, outcome, and intervention entity detection from CCS. CCS Explorer is packaged in a web-based graphical user interface where the user can provide any disease name. CCS Explorer then extracts and aggregates all relevant information from articles on PubMed based on the results of an automatically generated query produced on the back-end. For each task, CCS Explorer fine-tunes pre-trained language representation models based on transformers with additional layers. The models are evaluated using two publicly available datasets. CCS Explorer obtains a recall of 80.2%, AUC-ROC of 0.843, and an accuracy of 88.3% on sentence relevance prediction using BioBERT and achieves an average Micro F1-Score of 77.8% on Patient, Intervention, Outcome detection (PIO) using PubMedBERT. Thus, CCS Explorer can reliably extract relevant information to summarize articles, saving time by $\sim \text{660}\times$.
Motion recognition is a promising direction in computer vision, but the training of video classification models is much harder than images due to insufficient data and considerable parameters. To get around this, some works strive to explore multimodal cues from RGB-D data. Although improving motion recognition to some extent, these methods still face sub-optimal situations in the following aspects: (i) Data augmentation, i.e., the scale of the RGB-D datasets is still limited, and few efforts have been made to explore novel data augmentation strategies for videos; (ii) Optimization mechanism, i.e., the tightly space-time-entangled network structure brings more challenges to spatiotemporal information modeling; And (iii) cross-modal knowledge fusion, i.e., the high similarity between multimodal representations caused to insufficient late fusion. To alleviate these drawbacks, we propose to improve RGB-D-based motion recognition both from data and algorithm perspectives in this paper. In more detail, firstly, we introduce a novel video data augmentation method dubbed ShuffleMix, which acts as a supplement to MixUp, to provide additional temporal regularization for motion recognition. Secondly, a Unified Multimodal De-coupling and multi-stage Re-coupling framework, termed UMDR, is proposed for video representation learning. Finally, a novel cross-modal Complement Feature Catcher (CFCer) is explored to mine potential commonalities features in multimodal information as the auxiliary fusion stream, to improve the late fusion results. The seamless combination of these novel designs forms a robust spatiotemporal representation and achieves better performance than state-of-the-art methods on four public motion datasets. Specifically, UMDR achieves unprecedented improvements of +4.5% on the Chalearn IsoGD dataset.Our code is available at https://github.com/zhoubenjia/MotionRGBD-PAMI.
Multipath-based simultaneous localization and mapping (SLAM) is an emerging paradigm for accurate indoor localization with limited resources. The goal of multipath-based SLAM is to detect and localize radio reflective surfaces to support the estimation of time-varying positions of mobile agents. Radio reflective surfaces are typically represented by so-called virtual anchors (VAs), which are mirror images of base stations at the surfaces. In existing multipath-based SLAM methods, a VA is introduced for each propagation path, even if the goal is to map the reflective surfaces. The fact that not every reflective surface but every propagation path is modeled by a VA, complicates a consistent combination "fusion" of statistical information across multiple paths and base stations and thus limits the accuracy and mapping speed of existing multipath-based SLAM methods. In this paper, we introduce an improved statistical model and estimation method that enables data fusion for multipath-based SLAM by representing each surface by a single master virtual anchor (MVA). We further develop a particle-based sum-product algorithm (SPA) that performs probabilistic data association to compute marginal posterior distributions of MVA and agent positions efficiently. A key aspect of the proposed estimation method based on MVAs is to check the availability of single-bounce and double-bounce propagation paths at a specific agent position by means of ray-launching. The availability check is directly integrated into the statistical model by providing detection probabilities for probabilistic data association. Our numerical simulation results demonstrate significant improvements in estimation accuracy and mapping speed compared to state-of-the-art multipath-based SLAM methods.
Purpose: To optimize the choice of the flip-angles of magnetization-prepared gradient echo (MP-GRE) sequences for improved accuracy, precision, and speed of 3D-T1rho mapping. Methods: We propose a new optimization approach for finding variable flip-angle (VFA) values that improve MP-GRE sequences used for 3D-T1\r{ho} mapping. This new approach can simultaneously improve the accuracy and signal-to-noise ratio (SNR) while reducing filtering effects. We demonstrate the concept in the three different versions of the MP-GRE sequences that are typically used for 3D-T1rho mapping and evaluate their performance in model agarose phantoms (n=6) and healthy volunteers (n=5) for knee joint imaging. We also tested the optimization with sequence parameters targeting faster acquisitions. Results: Our results show that optimized VFA can improve the accuracy and the precision of the sequences, seen as a reduction of the mean of normalized absolute difference (MNAD) from 6~8% to 4~5% in model phantoms and from 14~22% to 12~14% in the knee joint, and improving SNR from 12~27 to 24~35 in agarose phantoms and 5~13 to 11~16 in healthy volunteers. The optimization can also compensate for the loss in quality caused by making the sequence faster. This results in sequence configurations that acquire nearly twice more data per unit of time with similar SNR and MNAD measurements as compared to its slower versions. Conclusion: The optimization of the VFA can be used to increase accuracy and precision, and improve the speed of the typical imaging sequences used for quantitative 3D-T1rho mapping of the knee joint.